Montreal, Quebec, Canada
Experienced mobile and full-stack software engineer working on large-scale software design projects for financial, government, and insurance organizations. Developed backend solutions using both Spring Boot apps and Node.js. On the mobile side my preferred technologies are multi-platforms frameworks like Flutter and React-native. Mobile & multi-platforms: - Flutter (Dart), React-native (Typescript), Android, iOS, Linux Webapp Front-End: - React, Typescript, Material-UI, AntDesign Fullstack with these BE: - Node.js, Java, Spring Boot, Kotlin, Groovy, AWS, GCP, Jenkins, SQL, noSQL, JMS
Develop a React Native mobile app, and also a Webapp RAG, integrating ONNX Runtime and Transformer.js to run advanced AI models offline, directly on iOS and Android devices. Designed for field workers in remote environments (e.g., mining), the app ensures data privacy, low latency, and cloud independence. 🔧 Technologies used: ONNX Runtime Mobile (optimized for CPU/GPU) Transformer.js for lightweight NLP WebAssembly (WASM) for browser-based inference Hugging Face models converted to ONNX 📌 Key Use Cases: Sentiment Analysis – Local BERT-based models analyze field reports in real time. Offline Translation – Whisper-style models enable multilingual communication without internet. Object Detection – YOLO/MobileNet models identify equipment or hazards via the device camera. Semantic Search (RAG) – Local embeddings and retrieval for querying technical documents. 🎯 Why Offline AI? Operates in connectivity-limited environments (e.g., underground galleries) Keeps sensitive data local (e.g., geological maps, maintenance logs) Reduces cloud costs and improves response time This project showcases how mobile-first AI can be deployed securely and efficiently in the field, even in the most challenging conditions.
Notable Project: In-Browser Semantic Search with LLMs Led the integration of a semantic search system within the open-source project coding-guide-helper, transforming it into a code review assistance tool. This project leveraged structured knowledge bases (coding guidelines) and enabled in-browser LLM-powered search using Transformers.js from Hugging Face: - Retrieved and processed Markdown files containing coding rules - Loaded a lightweight LLM model in-browser using Transformers.js - Optimized computed semantic vectors (embeddings) for each rule - Calculated embeddings for user queries and matched them to the most relevant rules Utilized ONNX Runtime with WebAssembly (WASM) to offload computation to the user’s CPU/GPU, reducing server dependency. This project demonstrates the potential of client-side LLMs to enhance developer experience and performance in a cost-efficient and scalable way. Also, as a mobile app developer, participate in the development and architecture of the native mobile app. - Technical leading & mentoring - Improve coding guidelines and practices - design UI (visual components) - implement business logic rules Environment: React 18.x, React Native, Hugging Face, ONNX Runtime, Transformer.js, Tanstack Query, React Hook Form, mmkv, geofencing, Axios, MobX, Typescript, GraphQL, Firebase, Okta, Gitub, Agile environment with scrums, stand-ups, and Jira
As a full-stack engineer, participate in the development and architecture of both native mobile and web applications using OAuth2 security pattern and material design principles for the user interface • Technical leading & mentoring • Improve coding guidelines and practices • React Native development with Paper • Webapp development with AntD Environment: React 18.x, AntD, Vite.js, React Native, Paper, OAuth2, AWS Cognito, React Query, Redux, Axios, Google Map, Git, Cypress, Python, Agile environment and Jira
As a full-stack engineer, participate in the development and architecture of the Beneva native mobile application using OAuth2 security pattern and material design principles for the user interface. • Flutter, React Native, Auth0, Firebase, Material Design, GraphQL, NestJS • Design UI (visual components) Environment: React 17.x, React Native, Paper, React Navigation, React Query, Axios, Typescript, Node.js, NestJS, Firebase: Authentication and Analytics, POC using Flutter and Auth0, Git, Agile environment with scrums, stand-ups, and Jira
As a full-stack engineer, participate in development of a React/Redux based data-visualization dashboard that customers use every day to guide how they spend their marketing budgets. Lots of cool, self-serve features that require robust technology. UI Environment: React, Redux, Redux-Saga, Typescript, Cypress, Cucumber, RTL, Jest, ChartJS, Material-UI Backend Environment (API): Groovy, Grails, Spring Boot, JPA, Security, WebClient, Hibernate, MySQL, Git, Docker, Nginx, AWS, S3, DynamoDB, Gradle, Sendgrid, Slack integration, Flyway, Handlebar, Google Script Management style: Agile, Scrums, Standups, Jira